AI Solution Architect, AI 解決方案架構師 (內湖瑞光)

Tasks

• Build LLM and VLM model serving platforms
• Conduct requirements interviews and feasibility assessments
• Deploy edge AI and AI server solutions
• Design enterprise AI architecture
• Design hybrid edge to cloud architectures
• Evaluate agent performance and model quality
• Handle cost monitoring token tracking and SLA planning
• Implement AI observability and governance
• Implement human-in-the-loop mechanisms
• Integrate AI agent platforms and enterprise systems
• Optimize inference performance and GPU utilization
• Orchestrate multi-agent workflows
• Plan PoC pilot and production rollouts
• Plan generative AI RAG and AI agent solutions
• Write technical proposals and architecture diagrams

Perks/Benefits

• N/A

Skills/Tech-stack

AI Agent | AI Foundry | AI Search | API Gateway | AWS Bedrock | Azure AI | Azure AI Foundry | Azure AI Search | Azure Kubernetes | Azure Kubernetes Service | Azure Machine Learning | Azure OpenAI | CRM Integration | CUDA | Cause analysis | Concurrency | Data Lake | Data Lake integration | Data Processing | Docker | ERP | Edge AI | Edge Computing | Embedding | Faiss | Generative AI | Google Vertex | Google Vertex AI | HTTP | HTTP API | Hybrid Cloud | Incident Response | Inference Server | JSON | JSON Web Token | KV cache | Kubernetes | Kubernetes Service | Language Model | Large Language Model | Latency | Logging | Machine Learning | Milvus | Monitoring | Multi-Modal | Multi-modal AI | NEMO | Nvidia Nim | OAuth | OAuth 2.0 | OAuth SSO | OCR | OCR Parsing | Observability | Pinecone | Prompt engineering | Python | Qdrant | RAG | REST | REST API | Root Cause Analysis | Root cause | SAP Integration | SGLang | SSO | Salesforce integration | Similarity Search | TensorRT | TensorRT-LLM | Throughput | Token Monitoring | Tool-Calling | Tracing | Triton Inference | Triton Inference Server | VLLM | Vector Database | Vector similarity | Vector similarity search | Vertex AI | Weaviate | Workflow automation